The most powerful tools of molecular and genetic medicine offer the ability to reliably detect sub-femtomolar (10-15) amounts of biomolecules obtained from tissues, e.g. polymerase chain reaction (PCR). This has resulted in a generation of diagnostic assays that have revolutionized medicine. Such assays for the early detection of bacterial sepsis have not been as successful as for viral infections due to the wide range of bacterial organisms that can lead to systemic disease. We have been investigating sensitive monitoring technologies that can be used to study bacterial infection and host response in living animal models of human disease, with the intent of applying these tools to the evaluation of human tissues for the presence of bacterial pathogens. Until recently obtaining spatiotemporal information about in vivo processes as they occur has been a fundamental difficulty in biomedical research. Molecular and cellular information has historically been obtained via ex vivo assays that are time consuming and conducted in the absence of contextual influences of intact organ systems. Understanding basic biological mechanisms requires that we are able to access this information in the living organism, in real time through imaging. Imaging strategies that employ molecular signatures as indicators of physiology and pathology can be used to reveal the location of infectious agents in vivo and the extent of the host response. The Contag lab has been exploring the use of biological light sources that report externally the inner workings of mammalian biology that can be built into animal models of human physiology and disease. Although mammalian tissues are relatively opaque, light is transmitted through tissues at low levels that can be externally monitored with sensitive detection systems. Whole body images of living animals expressing these optical tags reveal in noninvasive assays the relative number of labeled bacterial cells at specific sites or relative levels of gene expression. This method rapidly provides spatiotemporal information about interactive physiological processes in the context of whole biological systems, and thus enhances analyses of animal models. This approach rapidly provides in vivo biological data using tools that are easily accessible.